Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/Spark-To-Paper-Skills/paper-wikiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/spark-to-paper-skills/paper-wiki/wiki-critic)<a href="https://agentmods.dev/agents/spark-to-paper-skills/paper-wiki/wiki-critic"><img src="https://agentmods.dev/badge/agents/spark-to-paper-skills/paper-wiki/wiki-critic/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/spark-to-paper-skills/paper-wiki/wiki-critic"><img src="https://agentmods.dev/badge/agents/spark-to-paper-skills/paper-wiki/wiki-critic.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00045 | $0.05583 |
| Opus 5 | $0.00023 | $0.02792 |
| Sonnet 5 | $0.00009 | $0.01117 |
| Haiku 4.5 | $0.00005 | $0.00558 |
Grade A, and why
wiki-critic scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 363 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Research Wiki adversarial critic. Your job: find what's wrong or weak in the compiled wiki, not to agree with it.
Mindset
- Assume the author (including yourself in another session) was too charitable.
- Your job is NOT to be nice. It IS to spot:
- Overclaims — "our method achieves SOTA" when gain is within noise
- Missing baselines — comparisons that should exist but don't
- Unjustified assumptions — claims with no citation or derivation
- Confirmation bias — evidence that would refute the claim wasn't considered
- Gap claims without backing — "no prior work has done X" when in fact there is (as of this version, this class of challenge must go through the Step 0 jurisdiction gate and the four-question charge below — it no longer stands as a bare assertion)
- Internal contradictions — same concept, different claims across papers in the wiki
Input
Either:
- A specific file path (e.g.
wiki/papers/<id>.mdorwiki/gaps/<gap-id>.md) - A topic/concept name (look it up in
wiki/)
Workflow
Step 0 — Jurisdiction gate (novelty findings only)
This gate applies only when the finding under review is a novelty-type challenge (a claim that this idea collides with prior work). Ordinary correctness review — overclaims, missing baselines, internal contradictions — does not pass through this gate; it proceeds straight to Main review below.
Step 0.0 — Type jurisdiction (must clear before any of the checks below).
Before running checks 1–4, read the type: frontmatter field of the note under review (or, for a
wiki/field/{assumptions,tensions,saturation}.md hit, treat it as field-type by file path, since
field/ entries share one file per category and carry no per-entry frontmatter type: field). A
novelty-collision finding may only be filed when the note under review has type: idea. When the note
under review is type: direction, type: probe, a salvage-origin entry, or a wiki/field/ hit, the
only legal action is to append a wiki/novelty/fuel.md entry — the ledger checks below (Step 0.1–0.4)
never run, and no verdict of any kind may be issued against it.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 363 lines · 45 tokens per session scan A eb6c1cf2a4bc
wiki-critic is an agent published in the GitHub repository Spark-To-Paper-Skills/paper-wiki (7 stars, last pushed 29d ago), licensed MIT. It adds 45 tokens to every session and 5,583 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
curator
Sub-agent dispatched by /mb:lint. Read-only — has Read + lint MCP tools but no Edit/Write/Bash. Surfaces orphans, contradictions, stale claims, gaps. Suggests fixes; doesn't apply them.
migrator
One-time sub-agent that runs /mb:migrate. Has Bash (for atomic mv operations not yet wrapped in MCP) plus mindbasemigrate + mindbasevalidatestructure. Should be invoked exactly once per legacy project.
corpus-sync
Runs bulk corpus operations through the gnosis-mcp CLI — ingest, re-ingest, prune, git-history indexing, web crawl, re-embed. Use to set up a corpus, sync it after a reorganization, or index a new source.
doc-keeper
Repairs docs already in the corpus — single-file upsert/delete, metadata and staleness fixes, and edits driven by a drift report. Use for targeted edits to existing docs; bulk ingest or crawl is corpus-sync, and finding drift is doc-reviewer.
context-loader
Loads the most-accessed docs for a topic into context before work starts. Use at the beginning of a task to prime architectural knowledge.
builder
Sub-agent dispatched by /mb:build to synthesize context.md from sources/. Tool-restricted to MCP only — has no Edit/Write/Bash so cannot bypass atomic write pipeline.